News

Medical Transcription and EMRs: Opportunity Lost?
Posted by Claudia Tessier on September 12, 2008

Capturing patient information is a critical stage of the healthcare process. The methods chosen to do so can affect everything from the amount of time a physician spends with a patient to the accuracy of the health record created from the information collected and to the availability and value of that information to subsequent healthcare providers. As the healthcare community more universally incorporates the use of electronic medical records and moves toward e-care, the number and methods of information capture are expanding. These options stimulate debate about whether or not to transition to a new form of patient documentation – and if so, which one – or to continue to use traditional methods, most commonly medical transcription. This article will focus on medical transcription in relationship to electronic medical records (EMRs)

When considering information capture choices, practitioners and administrators must carefully consider not just current needs but also future goals. Depending on the size, style, and type of facility, requiring a single information capture method could result in additional costs or lead to savings of millions of dollars in a hospital. It could attract physicians or make a hospital unpopular with doctors. The typical conclusion is: Let doctors use any method they want to use, while offering/encouraging adoption of methods that facilitate point-of-care and real-time documentation. Traditional medical transcription continues to be the choice of many physicians, especially among those who have routinely been using this form of documentation throughout their career, but is it the best choice for healthcare? And if it is, for how long will it continue to be?

Nationwide, transcription is a huge industry. Medical Records Institute estimates that 90% of information capture is dictation and transcription, compared with less than 3% front end speech recognition and about 6% direct physician input by keyboard, stylus, touch screen and other methods. (New technologies such as digital pen have almost no impact at this time [1% or less].) While no hard figures exist on the size of the transcription industry, AHDI (Association for Healthcare Documentation Integrity – formerly AAMT, the American Association for Medical Transcription) reports that “current global medical transcription service expenditures are estimated between $12 billion and $20 billion annually, with the United States being the largest market.”¹ (Estimates in past years ranged as high as $28 billion for the US market alone.) The medical transcription industry, populated by a few, very large corporations, a wide range of small to large businesses and cottage industry outlets, and innumerable solo contractors, is a very private industry, one that has neither routinely nor widely shared statistical information about its operations and finances, which at least partly explains the great disparity and uncertainty in estimates. Further, it’s only in recent years that the healthcare industry in general has closely examined the costs of documentation, perhaps stimulated by the move toward EMRs and the concomitant increasing awareness that information capture is the key to EMR success.

Another way to estimate the size of the transcription industry is to survey the number of documents dictated and assess how many transcriptionists would be needed to transcribe these documents. Estimates several years ago indicated that the volume of reports traditionally dictated (including H&Ps, discharge summaries, consultations, operative reports, and radiology) would require somewhere between 300,000 and 400,000 medical transcriptionists (MTs) to transcribe them. This contrasts significantly with US Bureau of Labor Department statistics in 2006 that reported fewer than 100,000 employed MTs, with projected employment of 112,000 in 2016². Those figures are likely incomplete as the Bureau only recently began to count MTs as a separate category, and it will take time for respondents to change their reporting habits of including MTs among medical records clerks, medical secretaries, medical assistants, stenographers, and other categories of employees. If indeed these labor statistics are accurate, they emphasize a significant problem regarding the demand for dictation compared with the availability of employed MTs. The integration of back-end speech recognition with medical transcription is helping to meet this demand, as is offshore transcription, but the latter creates a range of reactions, from relief that there are more MTs somewhere – especially when offshore transcription charges less and offers overnight turnaround – to concerns for quality and security (the latter including both security of patient information and job security for MTs).

What is the Future of Medical Transcription?
Although visionaries of EMRs and EHRs (electronic health records) have been predicting the demise of medical transcription for almost 20 years, all have been proven wrong. While in the ambulatory sector, paperless EMRs have reached an estimated 15% to 20% level of adoption, hospitals are in the high nineties as they maintain transcription and paper parallel to electronic records. This duality represents a financial burden but is considered appropriate legal protection as many advisors do not consider the computer data integrity to be at the level of paper-based documentation, and further, some states do not yet recognize the legality of electronic documentation and/or signatures. As a result, many providers facilitate the creation of digital data on computers and then waste its potential benefits by printing, sorting and assembling the documents, then filing and maintaining them in traditional medical record charts. Further, they copy, scan, or fax such documents to make them available to various users. We are not aware that the costs of these duplicative, wasteful activities have been captured or reported, but estimates range from under $100 million dollars to billions of dollars, considering the thousands of hospitals and clinics and physician practices still adhering to these practices that perpetuate medical transcription as it was developed decades ago, even while concurrently adopting EMRs.

Transcription and EMRs
With the slow but steady move to electronic medical records, two questions about medical transcription emerge. First, when will direct data entry options have a significant effect on medical transcription, and second, what is medical transcription’s role in the transition to EMRs and computer-guided and computer-based care?

Certainly, medical transcription offers a bridge to EMR adoption, though historically, the overwhelming hope/desire among those administratively responsible for patient documentation has been that the EMR offers the best opportunity yet to get rid of transcription and its concomitant headaches – employment of a workforce that requires training and supervision and has high turnover as MTs move from job to job and also has huge costs, but that doesn’t consistently deliver in terms of quality and turnaround time. So, the budding EMR industry ignored the potential of transcription as a catalyst for EMR adoption, and the transcription industry was too fragmented and overconfident to recognize and respond quickly to the changes and threats that EMRs would bring. Had the transcription industry been willing and prepared early on to move toward integration with EMR and had the healthcare industry recognized the stimulus that medical transcription could bring to EMR adoption, they could have jointly benefited and advanced. Doing medical transcription electronically long preceded the first EMRs, so that electronic documentation would have been a natural and simplistic starting point around which to design EMRs. Unfortunately, that embryonic EMR revolution was aborted early on.

So, the question must be asked, can medical transcription still stimulate the EMR market? Well, yes and no. For those physicians who continue to choose medical transcription as their information capture method of choice, the transcription industry and EMR vendors should, even must, be responsive. What can be done? While health informatics professionals complain about the slow adoption of EMRs, the transcription industry and the EMR vendors should ramp up their cooperation to create uniform integration. Let every one of the 300+ EMR systems allow dictation and let the market determine whether the related turnaround time, quality, costs, etc. (see below) are acceptable. Let users dictate on cell phones and dictation devices, or through laptops and tablets—whatever their preference. And start addressing whether free text should be limited and templates developed that allow/push physicians to dictate in structured or semi-structured form. This too could have been initiated years ago. Why wait any longer?

The inherent concerns about medical transcription voiced by many providers, administrators, and executives remain and must also be addressed.
	Huge costs : Transcription costs per physician range from several thousand dollars to $25,000 and more annually. More and more physician practices and healthcare institutions are seeking alternatives by experimenting with back-end speech recognition, introducing front-end speech recognition, and using more point-of-care documentation devices and other technologies. While the success of such experiments is scattered, they are expected to become increasingly common.
	The demand for MTs exceeds the supply . Hospitals and clinics have trouble finding qualified transcriptionists. Rather than looking at the alternatives of new technologies, most are trying to solve their shortages by outsourcing transcription, which just shifts the point of impact. The integration of back-end and front-end speech recognition with medical transcription is helping to meet this demand, as is offshore medical transcription, but as noted above, resistance in some quarters remains strong. And, even with integrating those alternatives, the supply/demand problem remains.
	The shrinking MT population : “The MT profession is faced by the two-pronged problem of an ageing population without younger replacements.”3 The Bentley College study making this observation reports that 46% of its MT respondents are age 50 or more and 76% are age 40 or more, leaving fewer than 25% under 40. This ageing trend started years ago and shows no signs of reversing. The industry could simply die of old age.
	Turnaround time (TAT): Documentation of patient status and treatment should be immediately available to other healthcare professionals who need to base further decision making on previous documentation. If such information is not available due to transcription delays, the quality of care is diminished and patients could be harmed. In an ideal world, dictated information is immediately transcribed and available to other providers, but traditional (and back-end speech recognition-supported) medical transcription simply cannot meet the increasing expectations and value of this real-time documentation. A recent study of TAT by the American Health Information Management Association (AHIMA) and the Medical Transcription Industry Association (MTIA) reports “that very few standards for performance currently exist in the area of transcription TAT.”4 Noting that there is no precise TAT definition, the study chose what it identified as a widely held definition: “TAT for transcribed reports is the elapsed time from completion of dictation to the delivery of the transcribed document either in printed medium or electronically to a repository.”5 The AHIMA/MTIA study reported on separate surveys of HIMs (health information managers) and managers of MT/MTSO (medical transcription/medical transcription service organizations). The HIM survey reported that their contracted for and expected TATs for both paper and electronic documents range from 24 hours (for H&Ps, operative reports, consultations, progress notes, and pathology and cardiology reports) to 48 hours for paper discharge summaries and 48 to 72 hours for electronic output (!!). The MT/MTSO respondents reported similar TATs for paper documents, i.e., 21 to 40 hours for the same reports noted above, with radiology reported separately at 12 hours. When limited to electronic reports, the range improved to 18 to 35 hours, with radiology, again reported separately, having the best TAT at 10 hours.6 Of course, it should be acknowledged that many providers have digital dictation systems that allow access to the stored voice dictation during TAT, but this can create problems, particularly if the originator is difficult to understand, leading to the potential that different listeners may interpret and act on the dictation differently—with little or no documentation other than the voice file to back up what a particular listener heard. This means that, if the dictation is accessed by someone other than the originator and used even in part as the basis for patient care decisions and treatment, then that voice file must identify who accessed it when and must be preserved as part of the patient record, making it discoverable should there be legal action requiring that the patient’s record be produced. While the TATs reported in the AHIMA/MTIA study represent improvements over past transcription TAT performance (which sometimes extended to days, even weeks), they don’t meet the need (sometimes demand) for real-time, point-of-care documentation. Thus, alternative options that facilitate such documentation – front-end speech recognition, direct data entry via keyboard, pull-down lists, touch and click, etc. – are increasingly attractive. Some physicians are choosing to use scribes in exam rooms, presenting a career transition opportunity for MTs. As scribes, MTs (or others) do direct, real-time entry of patient information into EMRs as physicians “dictate” their findings and assessments during or immediately following patient encounters. While none of these alternatives represent an immediate threat to the medical transcription industry, it would be a mistake to dismiss them as unrealistic and not worthy of serious attention by the industry.
	Control and access : At least somewhat related to turnaround time are issues of control and access. Electronic dictation/transcription developments are partially addressing this problem, for example, by allowing access to voice files and transcription in process (which raises the concerns expressed in the previous section), but originators are still left with adjusting to the transcription industry’s limitations rather than having immediate access and control of their documentation, which real-time, point-of-care documentation options give them. MRI’s 2007 Survey of EMR Trends & Usages cited the following factors as most driving the need for EMR systems in medical practices
	» Improved patient documentation » Efficiency and convenience to physicians through workflow benefits » Remote access to patient information
	Transcription quality : Whatever the real numbers of MTs, the range of their abilities, like those for any profession, varies from poor to excellent, resulting in work products with a similar range of quality. Some years ago, when the author was CEO of the American Association for Medical Transcription—now the Association for Healthcare Documentation Integrity—a Hay Management Consultants study identified three levels of MTs, with level 3 MTs having the most expert depth and breadth of professional experience and serving as a resource to originators, other MTs, other healthcare providers, and students8 The author estimates that less than 20% of the MT population meets this level 3 definition. Supporting this view is the reality that, 25 years after the introduction of MT certification, there are fewer than 4000 CMTs (certified medical transcriptionists) even though eligibility for the certification exam (which is offered by AHDI) requires only two years of experience as an MT in an acute care setting. Further support is demonstrated in the recent statement from MTIA strongly advocating professional credentialing in response to the “need to promote the role of a highly skilled knowledge worker”9 in healthcare documentation and suggesting that when seeking level 2 transcriptionists, employers should indicate that certified status be required or preferred. (An RMT (registered medical transcriptionist) credential is recommended for entry level, or level 1, transcriptionists.) Notably, the MTIA statement does not speak to recruiting level 3 MTs, nor indeed is there a professional credential aimed at level 3 MTs. Finally, the quality of medical transcription has long been of concern in the healthcare industry, and indeed was a major factor behind the hope that EMRs would lead to better quality solutions. In fairness, however, quality deficits attributable to the originator who dictates patient documentation must also be acknowledged, including mumbling, rambling, excessive speed, background noises (music, bathroom, family, traffic, etc.), mispronunciations, wrong words, to name the most obvious and common.